►
From YouTube: Kubernetes Community Meeting 20190920
Description
We have PUBLIC and RECORDED weekly meetings every Thursday at 5pm UTC.
See this page for more information: https://github.com/kubernetes/community/blob/master/events/community-meeting.md
A
Hi
everyone
and
welcome
to
the
kubernetes
community
meeting
this
week.
I
am
your
host.
My
name
is
jonas
Rosland
I
work
at
VMware
and
I'm,
an
I'm,
an
open
source
community
manager
at
VMware,
so
I'm
part
of
the
sig
contrib
X,
together
with
a
bunch
of
other
folks
on
the
call
here
and
I'm
super
happy
to
have
you
all
here.
First
of
all,
I
want
to
make
sure
that
everyone
understands
that
this
is
a
community
meeting
and
will
be
posted
publicly
on
YouTube.
A
B
D
A
We've
got
Tim
pepper
on
the
call
here,
and
then
we
have
three
SIG's
who
will
be
giving
us
some
updates.
It
will
be
safe,
lug
provider
with
Chris
Hogue,
a
we've
got
a
cigar
Kotecha
with
a
Jase.
We
got
Sig
API
machinery
with
Daniel.
Then
we
have
a
bunch
of
announcements
regarding
an
election
update,
the
kubernetes
contributor
summit
and
the
contributor
survey
and
of
course,
the
weekly
shoutouts.
So
let's
dive
right
into
it
and
I'm
gonna
hand
this
over
to
Tim
pepper,
with
the
fig
relief
hi.
E
E
Basically,
where
we
are
will
be
thawing
Master.
The
expectation
is
on
Monday
so
that
113
can
commence,
which
is
exciting
there
and
then
Thursday.
The
27th
of
September
should
be
our
final
release,
then,
for
one
12.0
and
that's
basically,
it
I
did
also
note,
since
we
haven't
had
updates
on
the
the
patch
releases
for
prior
updates
just
be
kind
of
an
FYI,
the
one
nine
ten
and
eleven
dot
releases
from
the
last
few
weeks,
I
put
in
the
meeting
notes
there
and
that's
it
for
me.
A
A
B
Some
history
there
so
yeah,
so
I'm,
Jay,
singer,
Dumars
I
am
co-chair
of
architecture
and
I
work
at
Google.
And
if
you
go
to
the
meeting
minutes
for
the
community
meeting,
it's
that
Lee,
slash,
Cates
community
will
get
you
there
sure
not
there
already
and
basically
just
going
over
a
little
bit
about
what
architecture
does
and
what
we're
trying
to
accomplish
over
time,
which
is
primarily
to
manage
a
maintained,
architectural
consistency,
and
that
gives
us
sort
of
a
long
hall
view
of
what
design
decisions
are
in
communities.
B
And
how
do
we
maintain
a
logic
that,
if
you
walk
into
the
project
a
year
from
now,
it
makes
sense
to
somebody
who's
in
it
now
and
trying
to
to
make
sure
that
the
decisions
that
we
make-
and
you
know
that's
sort
of
the
sum
of
a
lot
of
small
decisions-
add
up
to
big
decisions.
So
that's
really
what
we're?
What
we're
watching
maintaining.
So
we
do
manage
a
few
sub
projects.
B
That's
a
really
great
service
and
people
who
are
interested
in
learning
more
about
api's
can
track
those
issues
and
those
issues
are
all
on
a
project
board,
and
that
is
how
we
manage
our
work
for
the
most
part,
so
do
not
seem
github
project
boards
and
are
familiar
with
Kanban
boards.
Essentially
it's
a
way
of
tracking
work
and
showing
two
things
that
there's
a
vertical
column
of
cards
and
essentially,
if
you're
doing
it,
if
you're
maintaining
the
board
regularly,
those
cards
should
be
stacked
in
rank
order.
B
So
basically,
the
things
at
the
top
are
the
things
that
are
going
to
move
to
the
next
state
the
fastest.
So
it
gives
you
to
a
really
quick
glance.
You
can
tell
what
the
prioritization
is
of
any
particular
phase
of
work
and
also
what
the
the
phases
of
the
process
are.
So
if
there's
say
our
review
phase
or
an
approval
phase,
those
things
should
all
be
represent.
This
columns
and
you'll
see
in
our
project
boards,
first
again
Sagarika
texture,
that
each
process
has
its
own
different
states.
B
B
Sig
architecture
is
really
responsible
for
keeping
an
eye
on
those
making
sure
that
as
people
introduce
new
functionality
into
the
ecosystem,
that
it
fits
with,
what
we're
trying
to
accomplish
textually
and
basically
just
providing
that
top
level
oversight
of
what
enhancements
are
being
introduced
into
the
project
and
also
that
the
project
has
a
same
processor
on
Kemps.
So
we're
really
trying
hard
to
find
ways
to
improve
that
and
make
it
more
consumable.
So
hopefully
there
will
be
some
updates
soon
about
some
enhancements
to
the
cap
process.
B
So
we'll
let
you
know
on
that
conformance
tests
or
another
thing
that
we
manage
conformance,
is
an
interesting
thing,
because
it's
really
the
concrete
definition
of
what
kubernetes
is.
We
talk
about
it
from
architectural
layers
standpoint,
but
really,
when
it
comes
down
to,
it,
can
forward,
says
sure
one
of
the
best
ways
to
consume
what
actually
is
the
functionality
that
should
be
provided
in
any
coronaries
class
trip.
So
so
those
are
the
sub
projects
we
manage.
B
We
also
have
sort
of
policy
and
governance
aspects,
so
I
mentioned
API
s,
and
so
we
do
have
API
governance,
so
we're
really
responsible
for
helping
to
maintain
the
guidance
and
review
process
for
api's
the
deprecation
policy
code
organization.
Generally
speaking,
the
kept
process
and
just
general
issues
around
criminality
scope
pretty
much.
B
If
you
go
to
the
agenda
the
day
before
so
basically
24
hours
before
the
meeting
and
there's
no
agenda
items,
that
means
that
the
meeting
won't
happen
so
we're
trying
to
prevent
needless
meetings
that
don't
have
a
gymnast
so
just
check
the
day
before
and
make
sure
that
there's
a
genetic
you
want
to
attend.
Otherwise,.
B
Yeah,
so
pretty
much
that's
it.
We
put
a
link
in
the
meeting
notes
for
this
meaning
to
you
can
look
at
our
agenda
and
meeting
notes
that
we
take
there
there's
a
link
to
that
and
also
a
link
to
our
YouTube
channel.
So
we
are
very
diligent
about
recording
and
posting
our
videos
of
our
meetings,
so
I
don't
believe
we
missed
any
in
the
last
year
or
so
so
I.
You
can
pretty
much
catch
up
to
everything.
That's
happening
at
any
given
moment
via
the
video
playback.
So
pretty
much
it
any
questions.
F
F
A
F
Okay,
so
yeah,
my
name
is
Chris
Hodge
and
I'm
with
sink
cloud
provider,
I'm
a
co-lead
of
it
with
Jago,
McLeod
and
Andrew
psych
him
we're
relatively
new
and
I
think
this
might
be
our
first
community
update.
Actually,
if
we
go
to
the
to
the
to
the
first
slide,
so
one
of
our
one
of
our
primary
goals
is
to
be
able
to
the
the
scope
of
the
sig
is
to
help
manage
all
of
the
individual
cloud
providers
in
producing
code
that
is
consistent
and
reliable
and
adheres
to
to
it
to
a
to
a
minimum
standard.
F
Part
of
the
reason
why
sinc
cloud
provider
arose
was
was
out
of
a
need
to
you
know.
We
were
you're
looking
at
a
number
of
cloud
providers
who
were
who
are
upstream
in
the
kerbin
IDs
codebase
right
now,
and
so
these
include
GCE
as
your
AWS
and
OpenStack,
but
we,
but
there
was
there,
there's
the
we
have
a
policy.
F
And
by
do
and
part
of
you
know,
we
we're
setting
requirements
that
you
know
minimum
set
of
requirements
to
add
new,
add
the
new
providers.
And
so
you
know,
these
include
things
like
making
sure
that
they're
documented
making
sure
that
they
are
posting
conformance
test
results
to
to
test
grid.
You
know
and
really
making
sure
that
when
users
come
and
they
find
a
cloud
when
they
find
it
provide
our
code
to
be,
you
know
they
can
use
of
urban
IDs
that
they
have.
F
You
know
a
minimum
set
of
testing
and
documentation
and
assurance
that
they're
going
to
have
a
positive
experience
out
of
it.
Additionally,
we've
started
some
initial
work
in
removing
the
entry
providers.
We've
formed
a
cloud
provider
extraction
working
group.
These
are
bi-weekly
meetings
every
third.
These
are
bi-weekly
meetings,
Thursday
at
11
a.m.
Pacific
time.
They
in
fact
we're
having
one
of
the
meetings
today.
If
anybody
is
interested
in
joining
that
and
we've
started,
that
Walter
is
doing,
you
know
a
huge
amount
of
the
work.
F
That's
that's
happening
on
this
and
it's
produced
a
you
know
a
document,
for
example
a
working
document
for
GCE
that
demonstrates
the
scope
of
work.
One
of
the
issues
that
we're
facing
with
the
entry
cloud
providers
is
to
varying
degrees.
They
have
used
internal
api's
and
so
api's
that
aren't
in
staging,
and
this
means
that
you
just
can't
pull
the
code
out
and
and
expect
it
to
work
without
potential
future
breakage,
because
internal
api's
aren't
guaranteed
to
exist.
F
F
You
know
pulling
the
code
out,
but
also
identifying
the
api's
that
need
to
be
moved
to
staging
so
that
so
that
the
providers
can
depend
upon
that
in
a
reliable
way
and
also
making
them
stop
dependent
depending
upon
the
internal
API
is
where
we
can't
move
like
staging,
and
so
it's
going
to
be
a
significant
amount
of
work.
And
if
we
move
on
to
the
next
slide
for
our
1.13
major
goals,.
F
So
we're
gonna
continue
this
work
to
remove
the
upstream
providers
so,
as
I
mentioned
before,
it
includes
identifying
internal
used
api's
and
we're
factoring
them.
It's
gonna
be
various
levels
of
work
amongst
the
different
providers,
but
the
short
term
solution
we
have
is
to
extract
everywhere,
it's
possible
and
vendor
current
IDs
within
the
provider
and
so
taking
care
of
this
internal
work.
F
You
know
the
one
of
the
primary
issues
there's
a
little
bit
of
a
chicken
and
egg
problem
where,
before,
when
you
shift
communities
and
there's
a
private
or
in
place,
you
can
just
flip
a
switch
and
it-
and
it's
enabled
now
you
have
to
load
a
control,
node
load,
the
control
plane,
you
know
on
a
provider
and
then
pull
the
provider
in
and
load
it.
You
know
and
load
it
in
and
so
there
may
be.
F
Other
things
are
also,
you
know,
ongoing
discussion
about.
You
know
considering
rehoming
all
the
providers
into
a
new
sync
cloud
provider
organization,
it's
probably
likely
far
beyond
the
scope
of
the
next
release
cycle,
and
is
you
know
really
something?
That's
a
much
larger
community
discussion
within
the
slides
here
we
have
a
link
to
that
discussion
on
the
sync
architecture.
Mailing
list-
and
you
know
kind
of
the
final
thing
that
we
need
to
do
and
what
they're
keen
is
is
continue
to
click
conformance
results
from
all
the
providers.
F
You
know
what
their
mandate
is.
It's
beyond
their
expertise.
You
know
they
shouldn't
be
expected
to
understand
how
all
these
providers
work
and
so
sick
cloud
provider
is
really
working
towards
again
building
a
standard
of
what
common
documentation
should
look
like
for
enabling
and
using
provider
code
and
what
individual
documentation
should
look
like
from
each
provider
and
the
minimum
standards
for
that,
as
well
as
we
started
working
with
a
sick
cluster
lifecycle
which
here's
a
very
similar
goal
where
cluster
lifecycle
is
interested
in,
how
do
you?
How
do
you
deploy
and
manage
a
kubernetes
cluster?
F
You
know
taking
advantage
of
the
different
resources
that
are
available
for
the
provider
solutions.
You
know,
so.
We've
been
working
with
them
on
making
sure
that
we
sync
up
on
creation
of
repositories
for
cloud
providers
as
well,
as
you
know,
providing
support
where
we
can
for
for
communicating
with
the
individual
providers
and
I
think
that
wraps
up
our
our
update,
I'm
happy
to
take
any
questions
about
the
work
that's
doing
or
how
to
get
involved.
A
F
This
is
actually
a
huge
discussion.
That's
happening
right
now,
I
mean
there's
a
there's
a
and
it
really
you
know,
boils
down
to
the
existential
question
of
what
is
kubernetes.
You
know
as
kubernetes
a
kernel
that
provides
a
minimum
set
of
functionality
that
that
lets.
You
do
a
number
of
things
and
then,
and
then
you
plug
into
that.
You
know
you
bring
your
own
things
to
plug
into
that,
to
to
hang
off
of
that
functionality.
F
And
you
know
it's
it's
hard
to
capture
the
fall,
you
know,
or
is
it
or
is
it
the
collection
of
API
is
that
we
interact
with
you
know
as
Josh
is
pointing
out
in
the
chat
here
and
that
affects
the
disposition
of
you
know
where
that
coach
live
and
who
should
who
should
have
responsibility
for
it?
You
know
I,
think
that
that
that
talking
about
that
in
the
community
meeting
is
you
know
we
like,
especially
in
an
update
like
this,
is
well
beyond
the
scope.
F
You
know
of
the
update
itself,
but
the
I
think
the
discussion
in
the
cigar
context.
Your
mailing
list
is
a
good
place
to
start
in
our
sink
cloud,
providing
meeting
yesterday,
which
we
hold
every
other
week,
and
so
it
was
an
on
week
this
week
you
can
also
review
some
of
the
discussion
that
we
had
there.
F
You
know,
because
I
think
there
are
a
lot
of
varying.
You
know
differing
opinions
on
it
and
there's
really
not.
That
decision
is
to
wear
things
where
things
are
hasn't
been
made.
Yet
for
the
time
being,
we're
bringing
in
new
cloud
provider
repositories
into
the
kubernetes
organization,
I
mean
that's
our
that's
our
current
plan
for
right
now,
you
know,
and
that,
and
that
decision
is
based
on
making
sure
that
right
now
there's
a
there.
F
There
is
an
imbalance
between
the
entry
providers
and
the
out
of
tree
providers,
and
we
want
to
make
sure
that
everybody
is
being
brought
to
the
same
level
playing
field,
and
you
know
that's
that's
the
solution
we
have
right
now.
We
want
to
make
sure
that
we're
not
sure
circuiting
any
of
those
decisions.
While
this
is
being
discussed,
I.
H
F
H
F
A
I
I
thought
of
a
bunch
of
things
to
say
while
I
was
sitting
here,
so
let's
see
how
many
I,
how
many
I
get
I
thought.
First,
I'd
start
off
with
what
what's
going
in
the
the
ongoing
release.
We
have
dry
run
in
alpha,
so
you
can
now
send
an
API
request
and
the
server
will,
instead
of
actuated,
that
API
request
tell
you
what
it
would
have
done
if
you'd
sent
it
for
real.
So
we
expect
this
to
help.
People
predict
the
outcome
of
their
other
actions.
It's
it's
not
like.
I
It's
not
like
a
contract
with
the
server
like
if
you,
if
you
do
a
dry
run,
to
create
it's
not
you're
not
going
to
get
the
same
like
UID
or
the
name.
If
you
use
the
generate
name
feature
but
but
you'll
be
able
to
test
your
your
API
to
see
like
what
does
it
look
like
when
it
runs
through
the
web
books
and
all
that
stuff?
So
that's
an
alpha.
I
I
You
can
add
more
versions
to
your
CDs,
but
each
version
must
be
like
there's
no
schema
change
allowed,
so
we
plan
to
add
a
schema
change
and
we
got
bogged
down
a
little
bit
arguing
about
how
the
API
should
work.
That's
pushed
off
to
the
next
release,
but
expect
that
in
the
next
release,
I
would
say-
and
that's
going
to,
let
you
register
webhook
to
do
it.
A
schema
transformation
for
you,
so
you'll
be
able
to
to
bring
old
clients
and
new
clients
for
your
series,
I
mean
I
would
say.
I
Our
eventual
goal
is
for
all
the
various
ways
of
API
definition
to
have
roughly
the
same
feature
set
so
like
whether
you're
choosing
aggregated
api's
or
C,
RDS
or
you're.
Choosing
that
based
on
do
you
want
to
write
a
custom,
storage
and
implementation
and
not,
unlike
you,
need
some
sub
resource
or
something
that
is
relatively
easy
for
us
to
do.
I
E
D
I
Watch
out,
I
thought
I'd
give
an
update
on
on
apply.
We've
been
working
for
a
while
to
move
apply
to
the
server
we're
doing
this
init
feature
branch,
Antoine,
Jenny
and
I,
or
the
people
have
been
working
on.
This
have
been
having
a
few
meetings
over
the
last
week
to
figure
out.
What's
our
what's
our
remaining
steps
to
exit
the
future
branch
and
get
this
merged
in
a
alpha
quality,
so
that
so
that
we
can,
like
you,
know,
do
two
stuff
in
master.
I
The
feature
branch
is
kind
of
nice
because
we
can
write
code
now
when
we're
frozen,
but
also
we
clearly,
we
have
to
get
out
of
the
feature
branch
before
people
will
actually
be
able
to
use
this.
So
look
for
some
more
complete
designs
from
us.
I
don't
know
when
we'll
get
those
out,
but
hopefully
soon,
maybe
maybe
by
next
week
and
then
I
just
thought
of
some
general
informational
things
about
sig
API
machinery
that
almost
always
bear
repeating.
We
don't
own
your
API.
I
You
can
ask
us
for
an
API
review
but
probably
have
opinions,
but
we're
not
the
API,
reviewers
I
know
it's
confusing
because
I'm
in
both
groups,
but
API
machinery
doesn't
own
most
api's
in
kubernetes.
It
is
a
little
confusing
because
we
do
own
some
api's,
so
the
the
like
I
would
say
the
metadata.
That's
in
every
object,
like
the
format
of
that
I.
Think
API
machinery
owns
that
because
it's
very
general
we
owned
the
C
or
D
API.
Obviously,
then
we
own
we
own
all
the
like
web
hook.
I
Interface
API
is
like
the
communication
to
dynamic
admission
web
hooks.
We
only
the
the
API
between
calling
it
an
API
may
be
generous,
but
we
own
the
communication
between
the
aggregator
and
all
aggregated,
aggregated
api's.
We
own
the
storage
interface.
It's
not
a
it's,
not
really
a
real
API
at
the
moment,
but
if
it
were.
I
The
communication
pathway
between
API
server
and
Etsy
D,
and
we
own
some
controllers
in
the
system
that
actually
general
things
of
kubernetes
so
like
the
garbage
collector,
the
namespace
controller,
I'm,
probably
forgetting
a
few
I
count
the
quota,
the
quota
controller
in
that
category,
because
that's
that's
sort
of
mostly
just
counting
counting
resources,
the
the
part
that's
specifically
about
adding
up
how
much
CPU
and
RAM
people
have
requested.
That's
pretty
Cooper,
Nettie's
specific
and
not
really
API
machinery.
I
So
so
I
don't
know,
maybe
that's
a
little
mixed
up
and
then
well
I
think
what
could
even
could
even
argue
that
we
only
are
the
are
back
API
and
and
that
obviously
the
auth
sig
owns
that
more
directly.
But
if,
if
you
were
to
think
about
kubernetes
in
terms
of
what
is
API
machinery
specific
and
what
is
kubernetes,
cluster
specific,
like
far
back
clearly
goes
with
the
API
machinery,
it's
much
more
general
than
like
pods
and
nodes.
I
We
own
that
and
I
even
extend
that
to
the
controller
manager
binary
because
I
don't
know,
I've
never
been
able
to
figure
out
who
owned
that
sorry
I
just
decided
that
maybe
sig
API
machinery
owns
that
okay,
so
hopefully
that
will
help
you
better
figure
out
whether
you
should
tag
sig
API
machinery
on
your
PR
I
know
we
are
at
the
top
of
the
list
when
you
look
for
SIG's,
but
that's
just
because
of
the
English
language.
It
doesn't
necessarily
mean
you
have
to
assign
it
to
us.
I
We
need
to
have
a
system
that
balances
like
making
sure
everybody
gets
their
fair
share.
That
still
recognizes
that
some
components
are
more
important
than
other
components
like.
If
you
start
your
scheduler,
that's
probably
not
good.
If
you
start,
if
you
starve
the
cubelets
reporting
the
node
status,
that's
even
worse,
because
then
the
node
controller
will
come
in
and
delete
some
pods
that
you
probably
didn't
want
deleted.
So
it's
going
to
be
it's
going
to
take
some
engineering
to
figure
out
how
to
do
that
safely.
I
I
There's
a
number
of
there's
a
number
of
things
that
have
that
same
problem.
Finally,
I
just
say
one
thing
about
our
signe
ting
and
agenda.
We've
been
just
echoed
what
Jay
said
a
few
minutes
ago.
If
it
comes
to
Wednesday
morning
and
we
find
that
our
agenda
document
is
empty,
then
we
will
probably
cancel
the
meeting.
So
it
really
helps
to
know
in
advance
what
topics
we're
going
to
discuss.
So
we
can
make
sure
the
right
people
are
there.
I
So
if
you
you
want
to,
if
you
want
to
have
an
item
discussed
like
there's
a
there's,
a
calendar
reminder
on
Tuesday,
if
you
sign
up
for
our
signaling
list,
that
tells
you
when
the
when
the
agenda
closes,
it's
very
generous.
It
goes
all
the
way
to
Tuesday
afternoon
so
make
sure
your
agenda
item
is
on
the
list
by
then
okay
I
think
that's
everything
I
want
to
talk
about.
So
thank
you.
Awesome.
A
Thank
you
so
much
Dan,
so
I
have
a
quick
PSA
public
service
announcement.
I've
heard
a
lot
of
talk
regarding
the
the
calendar
and
making
sure
that
you
have
all
the
meetings
on
your
calendars.
So
one
of
the
things
that
you
want
to
make
sure
you
do
is
look
at
the
notes
for
this
meeting.
We
have
some
instructions
for
the
calendar
there
and
that's
a
ribbon.
Ettus
community
calendar
make
sure
you
subscribe
that
calendar
don't
copy
over
items
because
they
will
fill
up
your
your
calendar
don't
copy
over
things.
A
Just
subscribe
and
I
put
some
screenshots
in
there
on
how
to
actually
do
that
as
well.
That
way,
you
have
a
separate
calendar
for
all
the
kubernetes
community
events
and
meetings,
so
you
can
easily
track
them
and
it
keeps
them
updated
as
well.
So
if
I
zoom
URL
or
something
else,
changes
or
time
changes,
you
will
automatically
have
an
updated
calendar.
So
that's
the
quick
PSA
here
and
next
up.
We
have
George
who
can
talk
about
the
election
update.
C
Everybody
I'll
go
as
fast
as
I
can
on
George
Castro
I'm,
one
of
your
election
officers,
along
with
Parrish
Pittman
and
Igor
de
Varenne
Suki,
both
who
could
not
attend
this
meeting.
The
first
thing
is
a
link
to
all
the
info.
You
need
I
sent
the
election
announcement
yesterday
to
kubernetes.
Do
you
want
to
click
through
the
link?
C
So
far
we
have
677
people
who
are
eligible
to
vote,
and
so
far
144
of
you
have
voted
if
you're
in
voters
dot
nd,
that's
the
file
with
your
github
handles
whether
that
says
whether
you
can
vote
and
you
have
not
yet
received
valid.
Please
send
us
a
mail
at
community
act
or
Nettie's
that
IO
we'll
send
you
a
ballot.
There's
some
guessing
that
happens
here,
because
we
don't
have
an
exact
mapping
of
you.
C
Github
handle
to
two
email
addresses,
so
usually
the
ones
that
we're
catching
our
people
who
might
have
recently
moved
jobs,
or
things
like
that.
So
by
now
you
should
have
received
a
ballot
from
the
system
and
if
you
have
not,
please
send
that
mail
and
we
will
process
that
as
quickly
as
possible
and
a
reminder.
The
election
ends
on
October
3rd
next,
the
kubernetes
contributor
summit
is
actually
happening.
This
is
a
day
before
cube
Con
in
Seattle
click.
C
Through
that
information
there
we'll
send
you
to
a
github
page,
that's
kind
of
talking
about
what
the
contributors
summit
is
gonna,
be
the
second
link.
There
is
one
I
need
help,
especially
from
those
of
you
that
are
planning
on
attending
Leeds
tech
needs
people
who
are
who
are
planning
on
attending
the
contributors
summit.
Last
time
we
kind
of
did
up
basically
a
huge
version
of
this
meeting
that
everyone's
doing
status
updates
this.
C
We
got
a
lot
of
feedback
that
people
want
to
hear
more
things
about
sustainability
as
a
project,
and
so
what
we're
gonna
do
is
in
the
morning
it's
kind
of
gonna
be
like
a
topic
one
to
many
discussion.
For
example,
sig
arch
is
gonna,
go
first
and
talk
about
sustainability
of
the
project
kind
of
overarching
terms,
as
opposed
to
just
giving
status
updates,
and
then
the
afternoon
will
be
an
unconference
we're
still
looking
for
topics
for
those.
So
if
you're
looking
to
attend
the
contributor
summit,
please
please
check
out
that
document.
C
There's
a
checkbox
when
you
register
for
Q
con.
That
will
say
whether
you
want
to
go
to
the
contributor
summit
or
not,
but
this
these
events
will
be
attending
limited
because
we're
expecting
something
like
8,000
people
keep
Khan
and
obviously
we
can't
have
them
all
come
to
the
contributor
summit.
So
please
register
early,
so
you
can
get
in
there
Josh.
You
want
to
say
anything
about
the
contributors,
social
for
Shanghai,
those
so.
D
D
However,
we
will
have
a
social
that
evening,
which
will
include
the
new
contributors
and
existing
contributors.
We
are
going
to
have
a
panel
run
by
pooja,
which
is
going
to
consist
of
Chinese
contributors,
both
in
China
and
living
abroad,
talking
about
contributing
to
kubernetes,
well
Chinese
I
as
a
discussion
of
thing.
So,
if
you're
going
to
Shanghai
and
you're
arriving
sometime
during
the
day
of
the
13th,
which
would
be
a
good
idea
if
you
are
going
to
Shanghai
or
even
earlier,
please
join
us
for
the
contributor
social
and.
C
One
thing
I
should
have
been
clear
on
the
contributor
summit
in
Seattle
is
going
to
be
in
two
parts:
those
of
you
that
are
10.
This
meeting
will
probably
go
to
the
existing
contributor
summit,
which
is
going
to
be
the
unconference
and
the
one
of
many
talks
I
talked
about,
but
for
people
who
are
brand,
you
do
the
project,
but
want
to
learn
how
to
contribute.
We
will
have
a.
C
We
will
have
a
separate
new
contributor
summit,
which
is
a
usual
Gwen
and
Josh
show
of
how
to
get
started
with
the
first
full
request:
how
to
sign
the
CLA
so
we're
gonna
try.
We
have
two
tracks
there,
one
for
existing
contributors
and
one
for
new
contributors.
If
you
have
any
questions
ping
us
in
the
SiC
contributor
experience,
slack
channel
and
we'll
be
happy
to
help
you
speaking
about
the
contributor
experience,
the
contributor
survey
is
out.
C
This
is
part
of
the
sig
contributor
experience
Charter
that
we
will
basically
ask
kubernetes
contributors
every
year
how
things
are
going,
how
you
answer
this
survey
as
a
contributor
will
basically
determine
the
things
that
sink
and
trim
backs
will
focus
on
over
the
next
year.
So
if
you
have
any
positive
bad
neutral,
whatever
experience
we're
looking
for
it.
Also,
please
click
through
that
survey.
Monkey
link
and
give
us
as
much
information
as
you
can.
That
would
be
really
great
Oktoberfest.
C
We're
sort
of
doing
this.
This
is
something
that
sponsored
a
digital
ocean
where
they
grab
new
people
who
want
to
get
started
in
open
source
and
ask
them
to
contribute
to
a
whole
bunch
of
open
source
projects.
I
am
putting
together
a
list
of
broken
404
links
in
some
of
our
properties
to
kind
of
give
them
a
place
to
go.
C
I
know
some
people
have
strong
opinions
about
having
people
doing
trivial,
pull
requests,
so
we're
gonna
try
to
intercept
that
by
giving
people
something
that's
easy
for
them
to
fix
and
without
hopefully
causing
too
much
havoc,
but
don't
worry,
I'm
working
with
the
github
org
admins
for
kubernetes
to
make
sure
that
we
don't
spam
everyone.
So
hopefully
that
will
go
great.
Any
questions
about
that
looks
like
no.
So
next
sig
leaves
we
put
together
some
recommendations
for
how
to
give
an
update
for
this
meeting.
So
some
people
say
hey
is
there?
C
Do
I
have
to
have
slides
or
not
what
should
I
be
talking
about,
so
we
went
ahead
and
pull
request
was
actually
a
while
ago.
We
just
forgot
to
tell
you
sorry
so
from
now
on
when
it's
your
turn
to
give
a
status
update
during
this.
During
this
meeting,
the
host
will
kind
of
remind
you
where
these
are
and
then
you
can
tailor
appropriately
and
last
call
for
outreach
e
in
turn.
Requests
I,
don't
know
who
added
this
bullet
if
they
want
to
talk,
but
that's
it
for
me
over
to
you
Jonas.
A
Right
so
less
topic
of
the
day
here
is
the
shout
out.
So
shout
out
is
the
thing
where
we
shout
out
to
people
that
have
done
amazing
things
in
the
kubernetes
community.
So
we
use
the
shout
outs
channel
on
the
kubernetes
lack
for
this.
So
if
you
have
any
chance
out
outs
to
any
individuals
or
team
members
that
you're
really
really
thankful
for
their
work,
make
sure
you
send
out
one
there.
So
the
first
one
is
from
Blanche
Schlossberg.
He
wants
to
give
a
huge
shout
out
to
George,
let's
see
here,
George
mr.
A
Next
one
is
from
Ben
the
elder.
He
wants
to
give
a
shout
out
to
mr.
hone
for
being
eternally
responsive
to
networking
issues
on
everything,
from
PR
reviews
to
cig
network
tests,
complex
DNS
images,
the
network
and
two
ends
and
answering
questions
related
to
network
issues
in
the
infra
and
helping
people
thanks
for
helping
to
helping
get
the
cube,
DNS
manifest
images
out
the
door
for
1.12
zeong
is
always
fixing
things
for
us
over
in
cig
testing
and
now
over
in
cig
release.
So
really
nice,
one
there
as
well.
A
The
the
next
one
is
from
Nikita
Regina.
She
wants
to
shout
out
to
Carolyn.
Let's
see
that
would
be
carolyn
van
flick
for
creating
lots
of
Help
Wanted
issues
on
the
server
catalog
Service
Catalog,
and
also
having
an
excellent
strategy
of
doing
that.
She
calls
the
pulse
at
the
strategy
where
she
put
out
a
bowl
of
mms
for
every
good
first
issue
that
she
creates.
She
needs
to
three
blue
M&Ms
and
she
goes
into
the
on
this
in
the
the
twitter
in
the
tweet.
A
That's
in
the
notes
there,
the
last
the
last
shout-out
here
is
from
devanam
Sreenivas,
who
wants
to
give
a
shout
out
to
Fabrizio
convenient
for
testing
all
the
things
version
1
to
12
are
C
1
cube
ad
cube
atom
under
various
scenarios.
So
big
shout
up
there
as
well
all
right
and
I
think
that's
it
for
today's
meeting.
Do.